A general purpose detailer, trained on images captioned both with Florence 2 Large rich captions and also WD-EVA02-Large-Tagger-V3 Booru tags. No particular trigger word. Recommended strength is around 0.6 to 0.8 or so, but feel free to adjust the strength as you see fit for any particular image.
Note that you CAN combine this Lora with my Zoot Fluxxer XL Lora if you wish, as both Loras were trained on completely different datasets.
Description
Expanded dataset by ~120 images. Should be a bit less oversaturated for photorealistic images now. Also trained this version for a bit longer than the initial one. Still no particular trigger word. All main showcase gallery images were still done just with base SDXL and this Lora.
FAQ
Comments (3)
I straight up meant to add naked dudes to the dataset to go along with the naked ladies for V2 but I forgot lol. I'll do it for V3.
Actually, i was just wondering what is new or fixed in V2 since the v1 was released a couple of days ago -- now i am at least aware what can we expect for V3, some release/version notes will be cool :)
It's a great LoRA, keep up the good work!
@diegocr I did mention some stuff in the about this version section